Pancreatic cancer is an almost uniformly fatal disease that will strike approximately 29,000 families in the United States this year. It has been estimated that 5-10 percent of these cancers have a familial basis. A better understanding of the genetic basis for the familial aggregation of pancreatic cancer will form a foundation for counseling patients and their families, and a basis for the rational application of new tests to detect this disease earlier. This project builds on progress made during the last funding period of this GI SPORE in which we demonstrated that germline mutations in the BRCA2, STK11/LKB1, hMLH1 and p16 genes are associated with an increased risk of developing pancreatic cancer; in which our segregation analysis of 273 pancreatic cancer cases and their families supported the involvement of a major gene in the familial aggregation of pancreatic cancer; and in which we prospectively demonstrated a significantly increased risk (18-fold, 95 percent CI 4.7-44.3) of pancreatic cancer in first-degree relatives of patients with familial pancreatic cancer. We now propose to: 1) determine the patterns of inheritance of pancreatic cancer and other cancers in families; 2) quantify the risk of pancreatic and non-pancreatic cancers in at-risk relatives; and 3) identify novel predisposing genes for the development of pancreatic cancer. The studies proposed here address a research priority identified in the NCI Pancreatic Cancer Program Review Group (PRG) - 'Identify genetic factors, environmental factors, and gene- environment interactions that contribute to pancreatic cancer development.' The studies proposed will utilize tissues and families from the Cores (Core 9002 and Core 9003) to translate genetic discoveries made in Project 0002 to patient care and they will provide a scientific basis for the selection of patients for screening using approaches developed in Projects 0013 and 0008.